Reproduce “Variational Inference MPC using Tsallis Divergence” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Ziyi Wang, Oswin So, Jason Gibson, Bogdan Vlahov, Manan S. Gandhi, Guan-Horng Liu, Evangelos A. Theodorou
Tsallis VI-MPC extends KL-based variational-inference control by using Tsallis divergence to weight control trajectories with a tunable cost cutoff and deformation parameter. Its weighting connects CEM’s hard selection with MPPI’s exponential weighting. Across five simulated systems and three policy classes, it had lower mean cost in all 15 pairings against both baselines, but variability was higher in some cases; real-world performance remains untested.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.